JP7114008B2 - data processor - Google Patents

Description

This disclosure relates to a data processing device and a data processing method.

Techniques for dividing time-series data into a plurality of partial time-series data are known.

For example, Patent Document 1 describes an extraction condition input unit that receives input of waveform data including change points of the state of the device, parameter information of the waveform data, and transition information of the device, time-series data of the device, and waveform data. Based on the similarity calculation unit that calculates the similarity with, the operation mode determination unit that sets the state of the device based on the transition information of the device, the calculated similarity, and the determined state of the device, A change point detection unit that detects a change point from the time series data of the device and sets the start time and end time of the segment that is a subsequence of the time series data, the state of the device, the start time of the segment, and the segment A data processing device is disclosed that includes an information output unit that outputs end time as segment information.
A data processing apparatus described in Patent Document 1 (hereinafter referred to as a "conventional data processing apparatus") extracts a waveform from time-series data using waveform data that serves as an index for division prepared in advance and parameter information of the waveform data. By detecting the pattern, the time-series data is divided into segments, which are partial time-series data, for each waveform data that appears repeatedly in the time-series data.

International Publication No. 2020/008533

In order to divide time-series data into segments, which are partial time-series data, a conventional data processing apparatus uses waveform data as an index for division and parameter information of the waveform data (hereinafter referred to as "special information"). There was a problem that it was necessary to prepare in advance.

This disclosure is intended to solve the above-mentioned problems, and divides the time-series data into partial time-series data for each waveform pattern that repeatedly appears in the time-series data without preparing special information in advance. It is an object of the present invention to provide a data processing device that enables

A data processing device according to this disclosure includes a time-series acquisition unit that acquires time-series data, a feature extraction unit that extracts a feature amount of the time-series data based on the time-series data acquired by the time-series acquisition unit, and a feature A first dividing unit that divides the time-series data into a plurality of first partial time-series data based on the feature amount extracted by the extracting unit; and a first dividing unit that divides the time-series data based on the feature amount extracted by the feature extracting unit. a second division unit for dividing each of the plurality of first partial time-series data into a plurality of second partial time-series data; and at least one of the plurality of second partial time-series data divided by the second division unit. an anomaly detection unit that detects an anomaly in the time-series data acquired by the time-series acquisition unit based on the second partial time-series data of the plurality of first partial time-series data divided by the first division unit; Each has first oscillation interval time-series data and first stable interval time-series data, and the second division unit divides each of the plurality of first partial time-series data divided by the first division unit into the first The first vibration interval time series data of the partial time series data is divided into a plurality of second partial time series data, and each of the plurality of second partial time series data divided by the second division unit is divided into the second oscillation interval Having time series data and second stable interval time series data, the anomaly detection unit, based on the second stable interval time series data among the plurality of second partial time series data divided by the second division unit, Detect anomalies in the time-series data acquired by the time-series acquisition unit .

According to this disclosure, time-series data can be divided into partial time-series data for each waveform pattern that repeatedly appears in the time-series data without preparing special information in advance.

FIG. 1 is a block diagram showing an example of a configuration of a main part of a data processing device according to Embodiment 1 and a data processing system to which the data processing device is applied. FIG. 2 shows an example of time-series data acquired by a time-series acquisition unit included in the data processing apparatus according to Embodiment 1, and first partial time-series data obtained by dividing the time-series data by a first division unit. It is an explanatory view showing . FIG. 3 shows the first partial time-series data obtained by dividing by the first dividing unit provided in the data processing device according to Embodiment 1, and the first partial time-series data obtained by dividing the first partial time-series data by the second dividing unit. FIG. 10 is an explanatory diagram showing an example of second partial time-series data; 4A and 4B are diagrams illustrating an example of a hardware configuration of main parts of the data processing apparatus according to Embodiment 1. FIG. 5 is a flowchart illustrating an example of processing of the data processing apparatus according to Embodiment 1. FIG.

Hereinafter, embodiments of the disclosure will be described in detail with reference to the drawings.

Embodiment 1.
A data processing device 100 according to a first embodiment and a data processing system 1 to which the data processing device 100 is applied will be described with reference to FIGS. 1 to 5. FIG.
FIG. 1 is a block diagram showing an example of a configuration of a data processing device 100 according to Embodiment 1 and a data processing system 1 to which the data processing device 100 is applied.
A data processing system 1 according to Embodiment 1 includes a storage device 10 , a display device 20 and a data processing device 100 .

The data processing device 100 acquires time-series data and performs predetermined data processing on the acquired time-series data. Details of the data processing device 100 will be described later.
The storage device 10 is a device that stores information necessary for the data processing device 100 to perform data processing. For example, the storage device 10 stores time-series data in advance.
The display device 20 is a device such as a display that displays and outputs a display image indicated by display information output by the data processing device 100 . That is, the data processing device 100 outputs the display information to the display device 20 and causes the display device 20 to display the display image indicated by the display information.

The data processing device 100 according to Embodiment 1 includes a time-series acquisition unit 110 , a feature extraction unit 120 , a first division unit 130 , a second division unit 140 , an anomaly detection unit 150 and an output unit 190 .
Note that the abnormality detection unit 150 is not an essential component in the data processing device 100 .
In Embodiment 1, the data processing device 100 is described as including the abnormality detection unit 150 .

The time series acquisition unit 110 acquires time series data.
The time-series data acquired by the time-series acquisition unit 110 is time-series information indicating physical quantities that are measured, measured, observed, aggregated, or the like at predetermined time intervals.
Specifically, the time-series data acquired by the time-series acquisition unit 110 is a time-series signal output from a sensor such as a vibration sensor, a distance sensor, a rotation sensor, a gyro sensor, a temperature sensor, or a sound sensor. It is converted into information. If the time-series data is time-series information indicating a physical quantity measured, measured, observed, or aggregated at predetermined time intervals, the signal output from the sensor is converted into time-series information. It is not limited to Note that the predetermined time interval does not have to be a uniform interval, and the time interval includes arbitrary intervals.

Specifically, for example, the time-series acquisition unit 110 acquires time-series data by reading time-series data stored in advance in the storage device 10 .
Since the time-series acquisition unit 110 may acquire time-series data, the source of the time-series data acquired by the time-series acquisition unit 110 or the time-series data acquisition unit 110 acquires the time-series data. The method is not limited.
Hereinafter, in Embodiment 1, the time-series data acquired by the time-series acquisition unit 110 will be described as being obtained by converting the signal output from the vibration sensor provided in the processing apparatus into time-series information. .

The feature extraction unit 120 extracts the feature amount of the time-series data based on the time-series data acquired by the time-series acquisition unit 110 .
Specifically, for example, the feature extraction unit 120 uses a sliding window to extract feature amounts from the time-series data acquired by the time-series acquisition unit 110 . In other words, the feature quantity output by the feature extraction unit 120 is information in which the values indicating the features of the regions in the time-series data corresponding to the sliding window are shown in time series.
More specifically, for example, the feature extraction unit 120 uses a sliding window to obtain a value obtained by subtracting the minimum value from the maximum value of the time-series data within the sliding window (hereinafter referred to as “amplitude value”) as the feature amount. Extract. The feature amount output by the feature extraction unit 120 is information indicating the amplitude values of the regions in the time series data corresponding to the slide window in time series.
Since the method of extracting feature values from time-series data using a sliding window is well known, the description thereof is omitted.

The first division unit 130 divides the time-series data into a plurality of first partial time-series data based on the feature amount extracted by the feature extraction unit 120 .
For example, the feature extraction unit 120 extracts the first feature amount from the time-series data acquired by the time-series acquisition unit 110 using a first sliding window having a predetermined first time length. The first division unit 130 divides the time-series data into a plurality of first partial time-series data based on the first feature amount extracted by the feature extraction unit 120 .
Specifically, for example, the first division unit 130 compares the amplitude value indicated by the first feature amount with a predetermined threshold value based on the first feature amount, and determines that the amplitude value equal to or less than the threshold value is larger than the threshold value. Identify the position in the first feature amount that increases. The first division unit 130 assumes that the end of the sliding window corresponding to the position is a change point of the time-series data (hereinafter referred to as "first change point"), and divides the time-series data at the first change point. by dividing the time-series data into a plurality of first partial time-series data.
For example, the first division unit 130 generates first division information indicating division positions in the time series data of each first partial time series data divided by the first division unit 130 .

In addition, the first dividing unit 130 identifies positions in the first feature quantity where amplitude values greater than the threshold are equal to or less than the threshold. The first division unit 130 assumes that the start time of the sliding window corresponding to the position is a change point of the time-series data (hereinafter referred to as "second change point"), and each first By dividing the partial time-series data, each of the plurality of first partial time-series data divided by the first division unit 130 can be divided into the first oscillation interval time-series data and the first stable interval time-series data. good.
When the first dividing unit 130 divides each of the plurality of first partial time-series data into the first oscillation interval time-series data and the first stable interval time-series data, the plurality of pieces divided by the first dividing unit 130 Each of the first partial time-series data has first oscillation interval time-series data and first stable interval time-series data.

In this case, for example, the first dividing unit 130 divides each first partial time-series data divided by the first dividing unit 130 into the time-series data of the first oscillation interval time-series data and the first stable interval time-series data. First division information indicating the position is generated.
In this case, the threshold is a predetermined value for dividing the first partial time-series data into the first vibration interval time-series data and the first stable interval time-series data, as described above. Therefore, for example, when the time-series data acquired by the time-series acquisition unit 110 is obtained by converting a signal output from a sensor such as a vibration sensor into time-series information, the threshold is set to 1 times the resolution of the sensor. , the first stable interval time series data can be extracted from the first partial time series data. Further, the same threshold value can be reused if the sensor or the device to be measured such as the processing device provided with the sensor has the same specifications. Therefore, the threshold can be easily determined.

The second division unit 140 divides each of the plurality of first partial time-series data divided by the first division unit 130 into a plurality of second partial time-series data based on the feature amount extracted by the feature extraction unit 120. .
For example, the feature extracting unit 120 uses a second sliding window with a predetermined second time length shorter than the first time length, from each of the plurality of first partial time series data divided by the first dividing unit 130 A second feature is extracted. The second division unit 140 divides the first partial time-series data into a plurality of second partial time-series data based on the extracted second feature amount.
Specifically, for example, the second division unit 140 compares the amplitude value indicated by the second feature amount with a predetermined threshold value based on the second feature amount, and determines that the amplitude value equal to or less than the threshold value is greater than the threshold value. Identify the position in the second feature amount that increases. The second division unit 140 assumes that the end of the sliding window corresponding to the position is the change point of the first partial time-series data (hereinafter referred to as the "third change point"), and at the third change point By dividing the time series data, the first partial time series data is divided into a plurality of second partial time series data.
For example, for each of the plurality of first partial time-series data, the second division unit 140 generates second division information indicating the division position in the time-series data of each second partial time-series data divided by the second division unit 140. Generate.

In addition, the second division unit 140 identifies positions in the second feature quantity where the amplitude value greater than the threshold is equal to or less than the threshold. Second dividing section 140 assumes that the start of the sliding window corresponding to the position is a change point of the time-series data (hereinafter referred to as “fourth change point”), and each second By dividing the partial time-series data, each of the plurality of second partial time-series data divided by the second dividing unit 140 can be divided into the second oscillation interval time-series data and the second stable interval time-series data. good.
When the second dividing unit 140 divides each of the plurality of second partial time-series data into the second oscillation interval time-series data and the second stable interval time-series data, the plurality of pieces divided by the second dividing unit 140 Each of the second partial time-series data has second oscillation interval time-series data and second stable interval time-series data.

In this case, for example, the second dividing unit 140 divides the second partial time-series data divided by the second dividing unit 140 into the time-series data of the second oscillation interval time-series data and the second stable interval time-series data. Generate second division information that indicates the position.
In this case, the threshold is a predetermined value for dividing the second partial time-series data into the second oscillation interval time-series data and the second stable interval time-series data, as described above. Therefore, for example, when the time-series data acquired by the time-series acquisition unit 110 is obtained by converting a signal output from a sensor such as a vibration sensor into time-series information, the threshold is set to 1 times the resolution of the sensor. , the second stable interval time series data can be extracted from the second partial time series data. Further, the same threshold value can be reused if the sensor or the device to be measured such as the processing device provided with the sensor has the same specifications. Therefore, the threshold can be easily determined.

2, time-series data acquired by time-series acquisition unit 110 included in data processing apparatus 100 according to Embodiment 1, and first time-series data obtained by first division unit 130 dividing the time-series data. Partial time series data will be explained.
FIG. 2 shows the time-series data acquired by the time-series acquisition unit 110 included in the data processing apparatus 100 according to Embodiment 1, and the first partial time-series obtained by dividing the time-series data by the first division unit 130. FIG. 4 is an explanatory diagram showing an example of data;

When the time-series data acquired by the time-series acquisition unit 110 is a signal output from a vibration sensor provided in a processing apparatus that repeatedly manufactures the same processed product converted into time-series information, the time-series In the data, a similar time-series value transition appears repeatedly each time a processed product is manufactured.
By dividing the time-series data into a plurality of first partial time-series data, the first dividing unit 130 divides the time-series data into a period during which each of the plurality of processed products is manufactured (hereinafter referred to as a "manufacturing period"). It can be divided into corresponding first partial time-series data.
The manufacturing period includes the processing period during which the workpieces to be processed are processed into processed products, and the idling period from the end of processing of a processed product to the start of processing of the next processed product. exists.
The first dividing unit 130 divides the first partial time-series data into the first vibration interval time-series data and the first stable interval time-series data, thereby dividing the first partial time-series data into the first time-series data corresponding to the processing period. It can be divided into one vibration section time-series data and the first stable section time-series data corresponding to the idling period.

Referring to FIG. 3, first partial time-series data obtained by division by first division unit 130 included in data processing apparatus 100 according to Embodiment 1, and second division unit 140 obtains the first partial time-series data. The second partial time-series data obtained by dividing the data will be described.
FIG. 3 shows first partial time-series data obtained by dividing by the first dividing unit 130 included in the data processing apparatus 100 according to Embodiment 1, and division of the first partial time-series data by the second dividing unit 140. FIG. 10 is an explanatory diagram showing an example of second partial time-series data obtained by

When the time-series data acquired by the time-series acquisition unit 110 is obtained by converting a signal output from a vibration sensor provided in a processing apparatus that repeatedly manufactures the same processed product into time-series information, the first In the partial time-series data, transitions of time-series values corresponding to each of a plurality of machining steps appear in one machining period.
Here, the processing process includes a process of placing a processed member on a pedestal, a process of detecting the shape of the processed member placed on the pedestal, a process of cutting the processed member placed on the pedestal, and a shape of the processed member after cutting. or a step of removing the processed member after cutting from the pedestal.

The second division unit 140 divides each of the first oscillation interval time series data in the plurality of first partial time series data into second partial time series data, thereby processing the first oscillation interval time series data into a plurality of processing. It can be divided into second partial time-series data corresponding to periods corresponding to each of the processes (hereinafter referred to as "process periods").
Each of the plurality of process periods includes a process operation period during which the processing equipment actually operates, and a period from when the operation of the processing equipment ends in one processing step to when the operation of the processing equipment starts in the next processing step. of process idling periods exist.
The second dividing unit 140 divides the second partial time-series data into the second vibration interval time-series data and the second stable interval time-series data, thereby dividing the second partial time-series data into the process operation period. It can be divided into the second vibration interval time-series data and the second stable interval time-series data corresponding to the process idling period.

With the above configuration, the data processing device 100, if the first time length and the second time length are predetermined, divides the time-series data into partial time-series data for each waveform pattern that repeatedly appears in the time-series data. It can be divided into series data.
In a conventional data processing device, a user analyzes time-series data acquired in the past, and obtains waveform data that serves as an index for dividing the time-series data, and parameter information of the waveform data (hereinafter referred to as "special information"). ) had to be prepared in advance. Analysis of time-series data requires considerable knowledge, and it is not easy for users to prepare special information in advance.

On the other hand, for example, the time-series data acquired by the time-series acquisition unit 110 is a signal output from a vibration sensor provided in a processing apparatus that repeatedly manufactures the same processed product, and is converted into time-series information. , the idle period and the process idle period are both known, so the user can easily determine the first time length and the second time length based on the idle period and the process idle period. can be done.
As a result, the data processing apparatus 100 can divide the time-series data into partial time-series data for each waveform pattern that repeatedly appears in the time-series data without preparing special information in advance.

The output unit 190 visualizes the first division information and the second division information based on the first division information generated by the first division unit 130 and the second division information generated by the second division unit 140, for example. A display image to be displayed is generated, and display information indicating the display image is output.
Specifically, for example, the output unit 190 outputs the display information to the display device 20 and causes the display device 20 to display the display image indicated by the display information.
The output unit 190 is not limited to outputting display information. For example, the output unit 190 may output the first division information and the second division information to the storage device 10 and cause the storage device 10 to store the first division information and the second division information.

The data processing device 100 may include the anomaly detector 150 as described above.
The abnormality detection unit 150 extracts the time-series data acquired by the time-series acquisition unit 110 based on at least one second partial time-series data among the plurality of second partial time-series data divided by the second division unit 140. Detect anomalies in For example, the anomaly detection unit 150 generates detection information indicating detection results.
Specifically, for example, an abnormality in the time-series data acquired by the time-series acquiring unit 110 based on the second stable interval time-series data among the plurality of second partial time-series data divided by the second dividing unit 140 to detect.

More specifically, for example, the anomaly detection unit 150 determines that the number of second stable interval time-series data corresponding to each of the plurality of first partial time-series data divided by the first division unit 130 is equal to all the first An abnormality in the time-series data is detected by determining whether or not the partial time-series data are the same.
If the number of second stable interval time series data corresponding to each of a plurality of first partial time series data is not the same for all first partial time series data, for example, the period between two processing steps may be shorter than the predetermined period. Also, in this case, it is conceivable that unexpected vibrations may occur in the processing equipment due to some cause during the period between the two processing steps.
Therefore, by configuring as described above, the abnormality detection unit 150 can accurately detect an abnormality in the time-series data.

Further, for example, the abnormality detection unit 150 determines that the second stable interval time-series data among the plurality of second partial time-series data divided by the second dividing unit 140 and the second stable interval time-series data are the first vibration Abnormalities in the time-series data acquired by the time-series acquiring unit 110 may be detected based on the second stable interval position information indicating the number of the second stable interval time-series data in the interval time-series data. .
For example, the anomaly detection unit 150 inputs the second stable interval time-series data and the second stable interval position information to a trained model prepared in advance. The anomaly detection unit 150 detects an anomaly in the time-series data based on the inference results output by the trained model. For example, the learned model is stored in the storage device 10 in advance, and the abnormality detection unit 150 acquires the learned model by reading the learned model from the storage device 10 .
By configuring as described above, the abnormality detection unit 150 can accurately detect an abnormality in the time-series data.

In addition, the learned model is learned by a learning device (not shown) based on the second stable interval time series data and the second stable interval position information. generated by Description of a method for generating a learned model is omitted.

Further, for example, based on the second stable interval time-series data and the second stable interval position information, the anomaly detection unit 150 divides the second stable interval time-series data into another first part by the second dividing unit 140. Second stable interval time-series data among a plurality of second partial time-series data obtained by dividing the first oscillation interval time-series data of the time-series data, the second stable interval having the same second stable interval position information An abnormality in the time-series data acquired by the time-series acquisition unit 110 may be detected by comparing it with the time-series data.
Specifically, for example, the anomaly detection unit 150 compares the second stable interval time-series data with other second stable interval time-series data having the same second stable interval position information, and determines the degree of similarity. By doing so, an abnormality in the time-series data acquired by the time-series acquisition unit 110 is detected. Since the method for determining the degree of similarity between time-series data is well known, the description thereof is omitted.
By configuring as described above, the abnormality detection unit 150 can accurately detect an abnormality in the time-series data.

The anomaly detection unit 150 may calculate the degree of anomaly of the time-series data when detecting an anomaly of the time-series data, and generate detection information indicating the calculated degree of anomaly.
The abnormality detection unit 150 calculates the degree of abnormality of the time-series data, so that the data processing device 100 can detect an abnormality such as deterioration of the processing device before an abnormality such as a failure of the processing device occurs. You can check the signs.

When the data processing device 100 includes the anomaly detection unit 150, the output unit 190 generates a display image that visualizes the detection information generated by the anomaly detection unit 150, in addition to the first division information and the second division information, for example. , outputs display information indicating a display image.
The output unit 190 outputs detection information to the storage device 10 in addition to the first division information and the second division information, and causes the storage device 10 to store the first division information, the second division information, and the detection information. There may be.

The hardware configuration of the main part of the data processing apparatus 100 according to the first embodiment will be described with reference to FIGS. 4A and 4B.
4A and 4B are diagrams showing an example of a hardware configuration of main parts of the data processing device 100 according to the first embodiment.

As shown in FIG. 4A, the data processing device 100 is composed of a computer, and the computer has a processor 401 and a memory 402 .
The memory 402 stores programs for causing the computer to function as the time series acquisition unit 110, the feature extraction unit 120, the first division unit 130, the second division unit 140, the abnormality detection unit 150, and the output unit 190. ing. The program stored in the memory 402 is read out and executed by the processor 401 to obtain the time-series acquisition unit 110, the feature extraction unit 120, the first division unit 130, the second division unit 140, the anomaly detection unit 150, and the output unit. 190 is realized.

Moreover, as shown in FIG. 4B, the data processing device 100 may be configured by a processing circuit 403 . In this case, the processing circuit 403 may implement the functions of the time-series acquisition unit 110 , the feature extraction unit 120 , the first division unit 130 , the second division unit 140 , the abnormality detection unit 150 , and the output unit 190 .

Also, the data processing device 100 may be configured by a processor 401, a memory 402, and a processing circuit 403 (not shown). In this case, some of the functions of the time-series acquisition unit 110, the feature extraction unit 120, the first division unit 130, the second division unit 140, the abnormality detection unit 150, and the output unit 190 are performed by the processor 401 and the memory 402. , and the remaining functions may be implemented by the processing circuit 403 .

The processor 401 uses, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a microprocessor, a microcontroller, or a DSP (Digital Signal Processor).

The memory 402 uses, for example, a semiconductor memory or a magnetic disk. More specifically, the memory 402 includes a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Programmable Read Only Memory), State Drive), HDD (Hard Disk Drive), or the like.

The processing circuit 403 is, for example, an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), an FPGA (Field-Programmable Gate Array), an SoC (System-on-a-Chip), or a system LSI (Large Integrated Circuit). ) is used.

The operation of the data processing device 100 according to the first embodiment will be described with reference to FIG.
FIG. 5 is a flowchart for explaining an example of processing of the data processing device 100 according to the first embodiment.

First, in step ST501, time series acquisition section 110 acquires time series data.
Next, in step ST502, feature extraction section 120 extracts a first feature amount.
Next, in step ST503, first dividing section 130 divides the time-series data into a plurality of first partial time-series data.
Next, in step ST504, feature extraction section 120 extracts a second feature amount.
Next, in step ST505, second dividing section 140 divides each of the plurality of first partial time series data divided by first dividing section 130 into a plurality of second partial time series data.
Next, in step ST506, abnormality detection section 150 detects an abnormality in the time-series data.
Next, in step ST507, the output section 190 outputs display information.
After step ST507, the data processing apparatus 100 terminates the processing of the flowchart.

As described above, the data processing apparatus 100 includes the time-series acquisition unit 110 that acquires time-series data, and the feature extraction unit that extracts the feature amount of the time-series data based on the time-series data acquired by the time-series acquisition unit 110. a first dividing unit that divides the time-series data into a plurality of first partial time-series data based on the feature amount extracted by the feature extraction unit 120; and a second dividing unit 140 that divides each of the plurality of first partial time-series data divided by the first dividing unit 130 into a plurality of second partial time-series data.
With this configuration, the data processing device 100 can divide the time-series data into partial time-series data for each waveform pattern that repeatedly appears in the time-series data without preparing special information in advance.

Further, in the data processing device 100 having the above configuration, the feature extraction unit 120 uses a first sliding window having a predetermined first time length to extract the first time series data from the time series data acquired by the time series acquisition unit 110 A feature amount is extracted, the first dividing unit 130 divides the time-series data into a plurality of first partial time-series data based on the first feature amount extracted by the feature extraction unit 120, and the feature extraction unit 120 divides the first partial time-series data into A second feature amount is extracted from each of the plurality of first partial time-series data divided by the first division unit 130 using a second sliding window having a predetermined second time length shorter than one hour length, The bifurcation unit 140 is configured to divide the first partial time-series data into a plurality of second partial time-series data based on the extracted second feature amount.
With this configuration, the data processing device 100 can divide the time-series data into partial time-series data for each waveform pattern that repeatedly appears in the time-series data without preparing special information in advance.

Further, in addition to the above configuration, data processing device 100 performs time series An anomaly detection unit 150 is provided for detecting an anomaly in the time-series data acquired by the acquisition unit 110 .
By configuring in this way, the data processing device 100 can accurately detect an abnormality in the time-series data.

In addition, in the above-described configuration, data processing device 100 divides each of the plurality of first partial time-series data divided by first dividing section 130 into first oscillation interval time-series data and first stable interval time-series data. For each of the plurality of first partial time-series data divided by the first division unit 130, the second dividing unit 140 divides the first oscillation interval time-series data of the first partial time-series data into a plurality of Each of the plurality of second partial time-series data divided into two partial time-series data and divided by the second dividing unit 140 has second oscillation interval time-series data and second stable interval time-series data, and is abnormal The detection unit 150 detects an abnormality in the time-series data acquired by the time-series acquisition unit 110 based on the second stable interval time-series data among the plurality of second partial time-series data divided by the second division unit 140. configured to
By configuring in this way, the data processing device 100 can accurately detect an abnormality in the time-series data.

Further, in the data processing device 100 having the above configuration, the anomaly detection unit 150 includes the second stable interval time-series data among the plurality of second partial time-series data divided by the second division unit 140, and the second The time series acquired by the time series acquiring unit 110 based on the second stable interval position information indicating the position of the second stable interval time series data in the first oscillation interval time series data. Configured to detect data anomalies.
By configuring in this way, the data processing device 100 can accurately detect an abnormality in the time-series data.

In the above-described configuration of the data processing device 100, the abnormality detection unit 150 converts the second stable interval time-series data to the second stable interval time-series data based on the second stable interval time-series data and the second stable interval position information. Second stable interval time-series data among a plurality of second partial time-series data obtained by dividing the first oscillation interval time-series data of other first partial time-series data by the dividing unit 140, the second stable interval position An abnormality in the time-series data acquired by the time-series acquisition unit 110 is detected by comparing it with the second stable interval time-series data having the same information.
By configuring in this way, the data processing device 100 can accurately detect an abnormality in the time-series data.

It should be noted that, within the scope of this disclosure, any component of the embodiment can be modified, or any component of the embodiment can be omitted.

A data processing apparatus according to this disclosure can be applied to a data processing system that detects anomalies in time-series data.

1 data processing system, 10 storage device, 20 display device, 100 data processing device, 110 time series acquisition unit, 120 feature extraction unit, 130 first division unit, 140 second division unit, 150 anomaly detection unit, 190 output unit, 401 processor, 402 memory, 403 processing circuit.

Claims (3)

a time-series acquisition unit that acquires time-series data;
A feature extraction unit that extracts a feature amount of the time series data based on the time series data acquired by the time series acquisition unit;
a first division unit that divides the time-series data into a plurality of first partial time-series data based on the feature quantity extracted by the feature extraction unit;
a second dividing unit that divides each of the plurality of first partial time-series data divided by the first dividing unit into a plurality of second partial time-series data based on the feature quantity extracted by the feature extracting unit; ,
Abnormalities in the time-series data acquired by the time-series acquiring unit are determined based on at least one second partial time-series data among the plurality of second partial time-series data divided by the second dividing unit. and an anomaly detection unit for detecting,
Each of the plurality of first partial time-series data divided by the first dividing unit has first oscillation interval time-series data and first stable interval time-series data,
For each of the plurality of first partial time-series data divided by the first division unit, the second dividing unit divides the first oscillation interval time-series data of the first partial time-series data into a plurality of Divide into second partial time series data,
Each of the plurality of second partial time-series data divided by the second dividing unit has second oscillation interval time-series data and second stable interval time-series data,
The anomaly detection unit obtains the time-series data obtained by the time-series obtaining unit based on the second stable interval time-series data among the plurality of second partial time-series data divided by the second dividing unit. to detect anomalies in
Data processing equipment. The abnormality detection unit divides the second stable interval time-series data among the plurality of second partial time-series data divided by the second dividing unit and the second stable interval time-series data into the first vibration interval. Detecting an abnormality in the time-series data acquired by the time-series acquisition unit based on second stable interval position information indicating what number the second stable interval time-series data is in the time-series data ,
2. A data processing apparatus according to claim 1 . Based on the second stable interval time-series data and the second stable interval position information, the anomaly detection unit detects the second stable interval time-series data as the second stable interval time series data among a plurality of the second partial time series data obtained by dividing the first oscillation interval time series data of the series data, wherein the second stable interval position information is the same Detecting anomalies in the time-series data acquired by the time-series acquisition unit by comparing with the second stable interval time-series data ,
3. A data processing apparatus according to claim 2 .

Images